Fine-grained Grading Research Articles

A fine-grained grading network for natural products based on dynamic association inference

Fine-grained grading of natural products, based on internal defects, plays an important role in food processing. Although X-ray imaging is critical for revealing internal defects due to its capacity to penetrate surfaces, the inherent grayscale and lower contrast make it challenging to extract fine-grained features. To address this, we propose the Fine-Grained Grading Network (FGGN) for classifying the quality grades of multiple natural products from X-ray images. Leveraging a ResNet18 backbone, FGGN enhances feature extraction by employing a hybrid attention module that emphasizes crucial local details by analyzing joint dependencies within tensor dimensions. Moreover, by combining cross-dimensional context encoding and dynamic graph convolution, a dynamic correlation inference module is proposed to construct dynamic graphs in a gated inference pattern, thereby integrating fine-grained information across abstraction levels. Finally, the cross-entropy loss is combined with graph constraint loss to train the network end-to-end. Extensive evaluation of three natural product datasets shows that the Fine-Grained Grading Network achieves state-of-the-art recognition performance, which demonstrates its grading applicability to multiple natural products.

Fully Automatic Fine-Grained Grading of Lumbar Intervertebral Disc Degeneration Using Regional Feature Recalibration Network.

Accurate fine-grained grading of lumbar intervertebral disc (LIVD) degeneration is essential for the diagnosis and treatment design of high-incidence low back pain. However, the grading accuracy is still challenged by lacking the fine-grained degenerative details, which is mainly due to the existing grading methods are easily dominated by the salient nucleus pulposus regions in LIVD, overlooking the inconspicuous degeneration changes of the surrounding structures. In this study, a novel regional feature recalibration network (RFRecNet) is proposed to achieve accurate and reliable LIVD degeneration grading. Detection transformer (DETR) is first utilized to detect all LIVDs and then input to the proposed RFRecNet for the fine-grained grading. To obtain sufficient features from both the salient nucleus pulposus and the surrounding regions, a regional cube-based feature boosting and suppression (RC-FBS) module is designed to adaptively recalibrate the feature extraction and utilization from the various regions in LIVD, and a feature diversification (FD) module is proposed to capture the complementary semantic information from the multi-scale features for the comprehensive fine-grained degeneration grading. Extensive experiments were conducted on a clinically collected dataset, which consists of 500 MR scans with a total of 10225 LIVDs. An average grading accuracy of 90.5%, specificity of 97.5%, sensitivity of 90.8%, and Cohen's kappa correlation coefficient of 0.876 are obtained, which indicate that the proposed framework is promising to provide doctors with reliable and consistent fine-grained quantitative evaluation results of the LIVD degeneration conditions for the optimal surgical plan design.

Multi-Task Learning for Diabetic Retinopathy Grading and Lesion Segmentation

Although deep learning for Diabetic Retinopathy (DR) screening has shown great success in achieving clinically acceptable accuracy for referable versus non-referable DR, there remains a need to provide more fine-grained grading of the DR severity level as well as automated segmentation of lesions (if any) in the retina images. We observe that the DR severity level of an image is dependent on the presence of different types of lesions and their prevalence. In this work, we adopt a multi-task learning approach to perform the DR grading and lesion segmentation tasks. In light of the lack of lesion segmentation mask ground-truths, we further propose a semi-supervised learning process to obtain the segmentation masks for the various datasets. Experiments results on publicly available datasets and a real world dataset obtained from population screening demonstrate the effectiveness of the multi-task solution over state-of-the-art networks.

Strength and reliability of WC-Co cemented carbides: Understanding microstructural effects on the basis of R-curve behavior and fractography

Strength and reliability of WC-Co cemented carbides (hardmetals) are dependent on effective fracture toughness as well as on nature, size and distribution of processing flaws. Regarding toughness, they exhibit a crack growth resistance (R-curve) behavior, derived from the development of a multiligament bridging zone at the crack wake. Accordingly, successful implementation of fracture mechanics requires consideration of tangency criterion, between applied stress intensity factor and R-curve, in addition to fractographic inspection. It is the aim of this study to evaluate the strength behavior of a series of experimental WC-Co grades on the basis of R-curve failure criteria. Results indicate that microstructural effects on the strength of hardmetals may be satisfactorily rationalized following the referred criterion. The analysis includes consideration of nature and distribution of fracture origins, found to be more diverse and wider, respectively, for the harder fine-grained grades. This experimental evidence, together with the fact that these hardmetals exhibit steeper rising R-curves than tougher coarse-grained ones, leads to lower reliability for the former. This investigation documents and validates the great relevance of R-curve behavior for optimizing the mechanical performance of WC-Co cemented carbides on the basis of microstructural design.

Fracture and fatigue of rock bit cemented carbides: Mechanics and mechanisms of crack growth resistance under monotonic and cyclic loading

In an attempt to improve the material selection, design and reliability of rock bit WC–Co cemented carbides (hardmetals), an extensive and detailed study is conducted with the main goal of characterizing the fracture and fatigue crack growth (FCG) behavior of four hardmetal grades. Work includes basic microstructural and mechanical characterization of the materials, assessment of fracture toughness and FCG kinetics. It is found that rock bit cemented carbides exhibit relatively high fracture toughness values (between 17 and 20MPa√m) in direct association with their specific microstructural characteristics, i.e. medium/coarse carbide grain size and medium cobalt content. The influence of microstructure on the measured crack growth mechanics under monotonic loading may be accounted by considering the effective operation of ductile ligament bridging and crack deflection as the prominent toughening mechanisms. Regarding FCG behavior, it is observed to exhibit a significant Kmax influence. Furthermore, relative increments in toughness are maintained, in terms of crack growth threshold, under cyclic loading. As a consequence, fatigue sensitivity for rock bit cemented carbides is found to be lower than that extrapolated from data reported for fine-grained grades. Crack growth resistance under cyclic loading for the hardmetals studied may be understood on the basis that prevalent toughening mechanisms (ductile ligament bridging and crack deflection) show distinct susceptibility to fatigue degradation and are thus critical in determining fatigue sensitivity.

Application of a profilometry method for the quantitative evaluation of fractures, secondary cracks and Palmqvist's cracks in sintered carbides

The paper presents the methodology and results of a quantitative description of fractures, secondary cracks and Palmqvist's cracks in sintered carbides by means of a profilometric method. It has been proved that the quantitative evaluation of decohesion mechanisms is possible thanks to a set of quantitative fractography parameters, which comprises fracture linear roughness, profile mean curvature, overlap coefficient and the WC phase dominance factor. Fractures of coarse-grained sintered carbides are characterized by a large profile development, while the fractures of fine-grained grades form fractures of a weaker development. A mean profile curvature gives one a quantitative evaluation of the share of transcrystalline cracks in relation to intercrystalline ones; coarse-grained grades form mixed brittle fractures—trans- and intercrystalline ones—whereas the fine-grained grades form an intercrystalline fracture only. The overlap coefficient describes the number of secondary cracks on a profile. Large, compact WC phase grains are conducive to the formation of cracks, which is a characteristic of coarse-grained grades. The value of the WC phase dominance factor is higher than 1, which proves that the crack initiation and propagation within the sintered carbides' structure is fostered by the WC phase. It was proved that the structural factor conducive to the propagation of Palmqvist's cracks is adjacent WC grains forming clusters.